1196

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 2, MAY 2004

Unit CommitmentA Bibliographical Survey

Narayana Prasad Padhy

AbstractWith the fast-paced changing technologies in the

power industry, new power references addressing new technologies are coming to the market. So there is an urgent need to keep
track of international experiences and activities taking place in
the field of modern unit-commitment (UC) problem. This paper
gives a bibliographical survey, mathematical formulations, and
general backgrounds of research and developments in the field of
UC problem for past 35 years based on more than 150 published
articles. The collected literature has been divided into many
sections, so that new researchers do not face any difficulty in
carrying out research in the area of next-generation UC problem
under both the regulated and deregulated power industry.
Index TermsAnt colony systems, artificial neural networks,
branch and bound, deregulation, dynamic programming, enumeration, evolutionary computation, expert system, fuzzy logic,
genetic algorithms, hybrid models, integer programming, interior
point, Lagrangian Relaxation, linear programming, priority list,
simulated annealing, tabu search, unit commitment.

I. INTRODUCTION

ANY utilities have daily load patterns which exhibit extreme variation between peak and offpeak hours because
people use less electricity on Saturday than on weekdays, less
on Sundays than on Saturdays, and at a lower rate between midnight and early morning than during the day [5], [13], [64], [66],
[104]. If sufficient generation to meet the peak is kept on line
throughout the day, it is possible that some of the units will be
operating near their minimum generating limit during the offpeak period. The problem confronting the system operator is to
determine which units should be taken offline and for how long.
In most of the interconnected power systems, the power
requirement is principally met by thermal power generation.
Several operating strategies are possible to meet the required
power demand, which varies from hour to hour over the day.
It is preferable to use an optimum or suboptimum operating
strategy based on economic criteria. In other words, an important criterion in power system operation is to meet the power
demand at minimum fuel cost using an optimal mix of different
power plants. Moreover, in order to supply high-quality electric
power to customers in a secured and economic manner, thermal
unit commitment (UC) is considered to be one of best available
options. It is thus recognized that the optimal UC of thermal
systems, which is the problem of determining the schedule
of generating units within a power system, subject to device
and operating constraints results in a great saving for electric
utilities. So the general objective of the UC problem is to
minimize system total operating cost while satisfying all of the
Manuscript received September 2, 2003.
The author is with the Department of Electrical Engineering at the Indian
Institute of Technology, Roorkee 247667, India.
Digital Object Identifier 10.1109/TPWRS.2003.821611

constraints so that a given security level can be met [61], [77],

[139].
This paper summarizes different methods used in the UC
problem-solving technique. It also presents a direction on
which the new solution techniques evolve with time. The
referred publications are extracted from the following journals
and proceedings:
IEEE TRANSACTIONS ON POWER SYSTEMS;
IEEE TRANSACTIONS ON POWER APPARATUS AND
SYSTEMS;
IEE Proceedings on Generation, Transmission and Distribution;
International Journal of Electrical Power and Energy
Systems;
International Journal of Electrical Power System Research;
International Journal of Power and Energy Systems;
International Journal of Electric Power Components and
Systems (earlier known as Electrical Machines and Power
Systems);
IEEE POWER SYSTEM REVIEW;
PROCEEDINGS OF IEEE POWER ENGINEERING SOCIETY
SUMMER/WINTER Meetings, PICA, etc.;
II. GENERAL BACKGROUND AND CONCEPTS
Various approaches have been developed to solve the optimal
UC problem. These approaches have ranged from highly complex and theoretically complicated methods to simple rule-ofthumb methods. The scope of operations scheduling problem
will vary strongly from utility to utility depending on their mix
of units and particular operating constraints [11], [35], [47],
[65], [72], [75].
The economic consequences of operation scheduling are very
important. Since fuel cost is a major cost component, reducing
the fuel cost by little as 0.5% can result in savings of millions
of dollars per year for large utilities [15], [110].
A very important task in the operation of a power system
concerns the optimal UC considering technical and economical
constraints over a long planning horizon up to one year. The
solution of the exact long-term UC [73], [118], [138] is not possible due to exorbitant computing time and, on the other hand,
the extrapolation of short-term UC to long-term period is inadequate because too many constraints are neglected such as maintenance time and price increases, etc.
Energy management systems have to perform more complicated and timely system control functions to operate a large
power system reliably and efficiently. In the case of a power
pool that involves several generation areas interconnected by
tie line [111], the objective is to achieve the most economical

0885-8950/04$20.00 2004 IEEE

PADHY: UNIT COMMITMENTA BIBLIOGRAPHICAL SURVEY

generation policy that could supply the local demands without

violating tie-line capacity constraints [16], [20], [76]. Although
the thermal and hydro thermal UC of a single area has been
studied extensively, the multiarea generation schedule has not
been given enough attention. The available literature for the UC
involving multiareas reveals that scheduling should be considered together with a viable economic dispatch to preserve the
tie-line constraints.
In the past, demand forecast advised power system operators of the amount of power that needed to be generated [117].
But under partially or fully deregulated environment, in the future, bilateral spot and forward contracts will make part of the
total demand known a priori [23]. The remaining part of the demand will be predicted as in the past. However, the generating
companies (GENCOs) share of this remaining demand may be
difficult to predict since it will depend on how its price compares to that of other suppliers. The GENCOs price will depend on the prediction of its share of this remaining demand
as that will determine how many units they have switched on.
The UC schedule directly affects the average cost and indirectly
the price, making it an essential input to any successful bidding
strategy. There may be a tendency to think that maximizing the
profit is essentially the same as minimizing the cost. This is not
necessarily the case. We have to remember that since we no
longer have the obligation to serve the demand, the GENCOs
may choose to generate less than the demand. This allows a
little more flexibility and makes the problem complex in the UC
schedules under the deregulated environment. Finally, the profit
depends, not only on the cost, but also on revenue. If revenue increases more than the cost does, the profit will increase. So for
the next-generation UC problem, researchers have to still play
an important role.
If the bid functions are nonconvex or nondifferentiable in nature, which is commonly seen in both regulated and deregulated
power industry, then the above problem becomes complex. Further, the complexity increases if the competition is encouraged
in both suppliers and buyers side including emission constraints.
So it has been observed that the hybrid models, which are the
combination of both classical and nonclassical methods, can
handle the present day complex UC problem commonly seen
within developed countries.
With the available standard software products, electric utilities have to enhance, evolve, and upgrade or add new applications such as UC solutions for modern deregulated power industry in conjunction with energy management systems [41],
[44], [85].

1197

are already exercising with the deregulated electricity industry.

Though there have been some pitfalls here and there, the end
users of the system are enjoying the fruits of the deregulated
electricity industry tree. So it is the high time for both the
developed and developing countries to modify or replace their
traditional algorithms based on the requirements of the modern
power industry.
In any restructured or deregulated power industry, the pool
implements a power action based on a UC model. Suppliers
submit their bids to supply the forecasted daily inelastic demand
[12]. Each bid consists of a cost function and a set of parameters that define the operative limits of the generating unit. After
the pool solves the UC problem, the system marginal price is
determined for each time period. The system marginal price is
nothing but the maximum average cost among the scheduled
generators. Several scheduling and pricing concerns have been
raised with the use of UC models to conduct power pool auctions [99], [127]. It is reported that the cost minimization model
does not always lead to lower prices when they are defined as
maximum average costs. Cost suboptimal solutions that result in
lower prices may exist and, therefore, the applicability of cost
minimization UC models for power pool auctions is questioned.
Chattopadhyay et al. [26] presented a model, capable of performing the following tasks: generation scheduling, interutility
transmission scheduling, and nonutility generation purchase
planning, etc. It is required to update the UC algorithm as
the electric industry restructures. In [23], a price/profit based
UC problem has been formulated which considers the softer
demand constraint and allocates fixed and transitional costs
to the scheduled hours. In August 2001, M. Madrigal et al.
[82] investigated the existence, determination, and effects of
competitive market equilibrium for UC power pool auctions
to avoid the conflict of interest and revenue deficiency. New
formulations to the UC problems suitable for an electric power
producer in an deregulated market has been provided by
Valenzuela et al. [60] and Larsen et al. [129] in 2001.
IV. UC PROBLEM FORMULATION
The generic unit commitment problem can be formulated as

(1.1)
where
(Fuel cost) is the input/output(I/O) curve that
is modeled with a curve (normally quadratic).

III. UC UNDER DEREGULATED POWER INDUSTRY

Since the mid-1980s, the electrical power-supply industry
around the world has experienced a period of rapid and critical
changes regarding the way electricity is generated, transmitted,
and distributed. The need for more efficiency in power production and delivery has led to privatization, restructuring, and,
finally, deregulation of the power sectors in several countries
traditionally under control of federal and state governments.
Many countries like England, the U.S., Canada, Australia, New
Zealand, Chile, Argentina, Peru, Colombia, and Scandinavian

(1.2)
, , and are the cost coefficients.
is described by
The maintenance cost
(1.3)
where
is the base maintenance cost, and
mental maintenance cost.
is described by
The startup cost

is the incre-

(1.4)

1198

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 2, MAY 2004

turbine startup cost;

boiler startup cost;
startup maintenance cost;
number of hours down;
boiler cool down coefficient.
Simillarly, the shut-down cost

New Power balance

(1.7)

is described by
(1.5)

where K is the incremental shut-down cost.

Subject to the following constraints:
Minimum up-time
a unit must be ON for a certain number of hours before
it can be shut off;
Minimum downtime
a unit must be OFF for a certain number of hours before
it can be brought online;
maximum and minimum output limits on generators

where
number of units;
number of time periods;
forecasted price for for period t;
up/down status of unit i;
power generation of unit i during time period t;
forecasted demand during time period t.
Reserve power and transmission losses are as per contract and
the rest of the constraints are the same as generic UC problem.
Similarly, network-constrained UC problem under both regulated and deregulated environment can be extended by incorporating the following system constraint parallel with (1.1) and
(1.6)
Power-flow equation of the power network

Ramp rate
where
Power balance
losses
Must run units
these units include prescheduled units which must be online, due to operating reliability and/or economic considerations;
Must out units
units which are on forced outages and maintenance are
unavailable for commitment;
Spinning reserve
spinning reserve requirements are necessary in the operation of a power system if load interruption is to be minimal. This necessity is due partly to certain outages of
equipment. Spinning reserve requirements may be specified in terms of excess megawatt capacity or some form of
reliability measures;
Crew constraints
certain plants may have limited crew size which prohibits the simultaneous starting up and/or shutting down
of two or more units at the same plant. Such constraints
would be specified by the times required to bring a unit
online and to shut down the unit.
Redefining the UC problem for the deregulated environment
[23] involves changing the demand constraints from an equality
to less than or equal, and changing the objective function
from cost minimization to profit (revenue-operational cost)
maximization. Now the generic UC problem under deregulated
environment can be formulated as

where
and

respectively calculated real and reactive

power for PQ bus i;
and
respectively specified real and reactive
power for PQ bus i;
and
respectively calculated and specified real
power for PV bus m;
and
voltage magnitude and phase angles at
different buses.
The inequality constraint on reactive power generation
at each PV bus

where
and
are, respectively, minimum and
maximum value of reactive power at PV bus .
The inequality constraint on voltage magnitude V of each
PQ bus

where
and
are, respectively, minimum and
maximum voltage at bus .
of voltage at
The inequality constraint on phase angle
all of the buses

where
and
are, respectively, minimum and
maximum voltage angles allowed at bus .
MVA flow limit on transmission line

(1.6)
Subject to the following constraints:

where
is the maximum rating of transmission
line connecting bus and .

PADHY: UNIT COMMITMENTA BIBLIOGRAPHICAL SURVEY

1199

V. METHODOLOGIES AND ANALYSIS

Kusic et al. [40] solved the UC of wholly owned and commonly

owned units. Hobbs et al. [135] developed and implemented a
realistic UC model in an energy management system.
The main aim of the UC solution is that it not only meets the
conventional requirements of having lowest fuel cost with spinning reserves and steady-state transmission capacity limits satisfied [19], but also offers good dynamic performance through the
enforcement of dynamic security constraints [141]. Li et al. [21]
introduced a new UC method based on decommitment procedure for solving the power resource scheduling problem. From
an initial schedule of all available units committed over the
study period, one at a time unit decommitment is accomplished
by dynamic programming. Sen et al. [127] solved the multiarea
UC problem using truncated dynamic programming and applied
to Indian Power System with two different areas having different
operating characteristics. Finally, Siu et al. [131] developed a
practical hydro dynamic UC and loading model for the British
Columbia Hydro Power Authority.

A. Exhaustive Enumeration
The UC problem has been earlier solved by enumerating all
possible combinations of the generating units and then the combinations that yield the least cost of operation are chosen as the
optimal solution. In [104], Kerr, et al., and in [66], Hara, et al.,
solved the UC problem successfully including Florida Power
Corporation by using the exhaustive enumeration method. Even
though the method was not suitable for a large size electric
utility, it was capable of providing an accurate solution.
B. Priority Listing
Priority listing method initially arranges the generating
units based on lowest operational cost characteristics. The
predetermined order is then used for UC such that the system
load is satisfied. Burns et al. [105] and Lee [38] handled the UC
problem, using priority order. Shoults, et al. [109] presented a
straightforward and computationally efficient algorithm using
priority order including import/export constraints. Lee [36] and
Lee et al. [34] solved the single and multiarea UC problem
using priority order based on a classical index.
C. Dynamic Programming
Stated in power system parlance, the essence of dynamic programming is for the total running cost of carrying x megawatt
(MW) of load on N generating units to be a minimum, the load
y MW carried by unit N must be such that the remaining load
of
MW is carried by the remaining
units also
at minimum cost. In mathematical form
where
minimum running cost of carrying x MW load
on N generating units;
cost of carrying y MW load on unit N;
minimum cost of carrying the remaining
MW load on the remaining
units.
Dynamic programming was the earliest optimization-based
method to be applied to the UC problem. It is used extensively
throughout the world. It has the advantage of being able to solve
problems of a variety of sizes and to be easily modified to model
characteristics of specific utilities [107], [136]. It is relatively
easy to add constraints that affect operations at an hour since
these constraints mainly affect the economic dispatch and solution method [33]. It is more difficult to include constraints [57]
that affect a single-units operation over time. The disadvantage
of the dynamic programming are its requirement to limit the
commitments considered at any hour and its suboptimal treatment of minimum up and downtime constraints and time-dependent startup costs [22].
In [98], Lowery discussed the practical applicability of
dynamic programming for UC solutions. In 1971, Happ [46]
reported the advantages of personal-computer solutions over
manual commitment solutions and claimed that the savings
obtained are in excess of 1% of the total fuel cost which
translates into U.S. $7000 for a 100-machine system. Pang et
al. [14] compared the performance of four UC methods, three
of which are based on the dynamic programming approach.

D. Integer and Linear Programming

Dillon et al. [130] developed an integer programming
method for practical size scheduling problem based on the
extension and modification of the branch-and-bound method.
The UC problem can be partitioned into a nonlinear economic
dispatch problem and a pure integer nonlinear UC problem
based on benders approach. Whereas the mixed integer programming approach solves the UC problem by reducing the
solution search space through rejecting infeasible subsets.
A linear programming UC problem can be solved either by
decomposing the whole problem into subproblems with help
of DantzigWolfe decomposition principle and then each
subproblem is solved using linear programming or the problem
can be solved directly by revised simplex technique [44].
E. Branch and Bound
Lauer et al. [42] and Cohen et al. [10] presented a new
approach for solving UC problem based on branch-and-bound
method, which incorporates all time-dependent constraints and
does not require a priority ordering of units. In [74], Huang
et al. proposed a constraint logic programming along with
the branch-and-bound technique to provide an efficient and
flexible approach to the UC problem.
The branch-and-bound procedure consists of the repeated application of the following steps. First, that portion of the solution space (i.e., set of decision variables under consideration)
in which the optimal solution is known to lie is partitioned into
subsets. Second, if all of the elements in a subset violate the constraints of the minimization problem, then that subset is eliminated from further consideration (fathomed). Third, an upper
bound on the minimum value of the objective function is computed. Finally, lower bounds are computed on the value of the
objective function when the decision variables are constrained
to lie in each subset still under consideration. A subset is then
fathomed if its lower bound exceeds the upper bound of the minimization problem, since the optimal decision variable cannot
lie in that subset. Convergence takes place when only one subset
of decision variables remains, and the upper and lower bounds
are equal for that subset.

1200

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 2, MAY 2004

F. Lagrangian Relaxation
Based on Lagrangian Relaxation approach, the UC problem
can be written in terms of 1) a cost function that is the sum terms
each involving a single unit, 2) a set of constraints involving
a single unit, and 3) a set of coupling constraints (the generation and reserve constraints), one for each hour in the study
period, involving all of the units. Formally, we can write the UC
problem as follows [30], [86], [126], [128]:
Minimize
Subject to the unit constraints
For all units I, where
and
; and the coupling generation and reserve
constraints

For all times t and requirement n. Everett showed that an

approximate solution to this problem can be obtained by
adjoining the coupling constraints onto the cost using Lagrange
multipliers. The resulting relaxed problem is to minimize the
so-called Lagrangian subject to the unit constraints

where
are the multipliers associated with the th requirement for time t. Describing the Lagrangian Relaxation method
requires answering the following questions: 1) how do we find
so that the solution to the relaxed problem
the multipliers
is near the optimum, 2) how close to the optimum is the solution, and 3) how do we solve the relaxed problem? Everett and
dual theory together provides us the insight to the above questions and solutions.
Lagrangian Relaxation is also being used regularly by some
utilities [9], [106], [112]. Its utilization in production UC programs is much more recent than the dynamic programming.
It is much more beneficial for utilities with a large number
of units since the degree of suboptimality goes to zero as the
number of units increases. It has also the advantage of being
easily modified to model characteristics of specific utilities. It
is relatively easy to add unit constraints. The main disadvantage
of Lagrangian Relaxation is its inherent suboptimality.
In [7], Merlin et al. proposed a new method for UC using
Lagrangian Relaxation method and validated at Electricite De
France. Aoki et al. [62], [63] applied Lagrangian Relaxation
method for a large-scale optimal UC problem, which includes
three types of units such as usual thermal units, fuel-constrained
thermal units, and pumped storage hydro units. A three-phase
new Lagrangian Relaxation algorithm for UC is proposed by
Zhuang et al. [32]. In the first phase, the Lagrangian dual of
the UC is maximized with standard subgradient technique,
the second phase finds a reserve feasible dual solution, and
followed by a third phase of economic dispatch. Wang et al.
[17] presented a rigorous mathematical method for dealing
with the ramp rate limits in UC and the rotor fatigue effect

in economic scheduling. Ma et al. [52] incorporated optimal

power flow in the UC formulation. Using benders decomposition, the formulation is decomposed into a master problem and
a subproblem. The master problem solves UC with prevailing
constraints, except transmission security and voltage constraints, by augmented Lagrangian Relaxation. The refinement
or reduction in complexity of Lagrangian Relaxation method
has also been suggested by Takriti [120] and Cheng [24].
G. Interior Point Optimization
Interior point methods have not only been used to successfully solve very large linear and nonlinear programming
problems, but also to solve combinatorial and nondifferentiable
problems. The interior point method has now been applied to
solve scheduling problems in electric power systems. Madrigal
et al. [81] applied the interior point method for solving the UC
problem, based on his observation of the interior-point method
to have two main advantages such as they have better convergence characteristics and they do not suffer with parameter
tuning.
H. Tabu Search
Tabu search is a powerful optimization procedure that has
been successfully applied to a number of combinatorial optimization problems. It has the ability to avoid entrapment in
local minima by employing a flexible memory system [1]. Mori
et al. [48], [149] presented an algorithm, embedding the priority list into tabu search for unit commitment. Rajan et al. [148]
solved UC problem using neural-based tabu search method. In
[150], Lin et al. developed an improved tabu search algorithm
for solving economic dispatch problems. Mantawy et al. [1],
[3] presented UC solutions using tabu search and also solved
long-term hydroscheduling problems very efficiently using a
new tabu search algorithm [151].
I. Simulated Annealing
Simulated annealing was independently introduced by
Kirkpatrick, Gela, and Vecchi in 1982 and Cerny in 1985.
Annealing, physically, refers to the process of heating up a
solid to a high temperature followed by slow cooling achieved
by decreasing the temperature of the environment in steps [68],
[133]. By making an analogy between the annealing process
and the optimization problem, a great class of combinatorial
optimization problems can be solved following the same
procedure of transition from equilibrium state to another,
reaching minimum energy of the system. In solving the UC
problem, two types of variables need to be determined. The
unit status (binary) variable U and V and the units output power
(continuous) variables, P. The problem can be decomposed into
two subproblems, a combinatorial optimization problem in U
and V and a nonlinear optimization problem in P. So, simulated
annealing can be suitably used to solve the UC problem.
Mantawy et al. [2] presented a simulated annealing algorithm
to solve the UC problem and concluded that even though simulated annealing algorithm has the disadvantage of taking long
CPU time, it has other strong features like being independent
of the initial solution and mathematical complexity.

PADHY: UNIT COMMITMENTA BIBLIOGRAPHICAL SURVEY

J. Expert Systems
Experience is knowledge and understanding knowledge
is intelligence. (Narayana Prasad Padhy)
Expert system is an intelligent computer program that uses
knowledge and inference procedures to solve problems that
are difficult enough to require significant human expertise
for their solutions [29]. Generally, its knowledge is extracted
from human experts in the domain and it attempts to emulate
their methodology and performance [67], [89]. As with skilled
humans, their knowledge in both theoretical and practical, have
been perfected through experience in the domain. Unlike a
human being, however, current programs cannot learn from
their own experience; their knowledge must be extracted from
humans and encoded in a formal language [37].
Expert system-based approach to short-term UC, which is intended to process large generating schedules in real time, [138],
[145], which combines database management with the latest developments in expert system design, and extensive use of manmachine interfaces are used for this investigation. The real-time
processing capability of the proposed system is challenged with
regard to those (integer and linear programming, Lagrangian
Relaxation, etc.) of mathematical programming methods [113],
[137].
Mokhtari et al. [114] developed an expert system-based consultant to assist power system operators in scheduling the operation of generating units. Ouyang et al. [146] proposed a UC
expert system consisting of a commitment schedule database, a
dynamic load pattern matching process, and an interface optimization process. Tong et al. [119] proposed an algorithm for
UC, uses priority list-based heuristics in the form of interface
rules to find a suboptimal schedule for a given load pattern. An
expert system was developed in 1991 by Salam et al. [83] and
used as a preprocessor as well as a postprocessor to the truncated
dynamic programming-based UC program to obtain an operationally feasible solution.
K. Fuzzy Systems
As complexity rises, precise statements lose meaning
and meaningful statements lose precisions. (Lotfi A.
Zadeh)
Zadeh is credited with introducing the concept of fuzzy sets
in 1965 as a mathematical means of describing vagueness in linguistics. The idea may be considered as a generalization of classical set theory. In the decade since Zadehs pioneering paper on
fuzzy sets, many theoretical developments in fuzzy logic took
place in the U.S., Europe, and Japan. From the mid-1970s to
the present, however, Japanese researchers have done an excellent job of advancing the practical implementation of the
theory; they have been a primary force in commercializing this
technology.
UC is a complex decision-making process [84] which operates appropriate units at different hours and schedules the outputs of the committed units to meet a predicted demand, such
that the operating cost is minimized. Due to the uncertainty of
the demand and outages of generating units [65], this raises the
question of how to tackle the UC problem when the load demand and other variables are imprecise [79], [88], [93], [94]. Re-

1201

searchers have observed that stochastic models perform better

than deterministic models under uncertainty but they have their
own limitations [121].
Tong et al. [119] provided a rational model by considering
the outage of thermal units and the uncertainty of the demand.
Zhai et al. [28] described a method for analyzing the effect of
load uncertainty on UC risk, which is the probability of having
insufficient commitment capacity to compensate for unit failures and/or unanticipated load variation. Saneifard et al. [122]
demonstrated the application of fuzzy logic to the UC problem.
This method allows a qualitative description of the behavior of
a system, the systems characteristic, and response without the
need for exact mathematical formulations.
L. Artificial Neural Networks
The mind being, as I have declared, furnished with a
great number of the simple ideas conveyed in by the senses,
as they are found in exterior things, or by reflection on its
own operations, takes notice, also, that a certain number of
these simple ideas go constantly together
which, by inadvertency, we are apt afterward to talk of and consider as
one simple idea. (John Locke, Essay Concerning Human
Understanding)
Artificial neural networks (ANNs) are intended to model the
behavior of biological neural networks. The original hope for
the development of ANNs is intended to take advantage of parallel processors computing than traditional serial computation.
Over the years, several models of ANNs and the learning algorithms that are associated with networks have been developed
[50].
Sasaki et al. [49] explored the possibility of solving the combinatorial optimization problem, in particular to UC applying
the Hopfield neural network. The proposed neural network has
solved a UC of 30 units over 24 periods, and the obtained results
are very encouraging. C. Wang et al. [18] proposed an ANN
model for UC with ramp-rate constraints. It has been found that
UC problem cannot be handled accurately within the framework
of the conventional Hopfield network. So, Walsh et al. [80] presented augmented network architecture with a new form of interconnection between neurons, giving a more general energy
function containing both discrete and continuous terms. Liang
et al. [108] successfully solved the UC problem using an extended mean field annealing neural network approach.
M. Genetic Algorithms
Again I saw that under the sun the race is not to the swift,
nor the battle to the strong, nor bread to the wise, nor riches
to the intelligent, nor favor to the man of skill but time and
chance happen to them all. (The Bible Ecclestastes, 9)
During the last 30 years, there has been a growing interest
in problem-solving systems based on the principles of evolution and machine learning [116], [134]: such systems maintain a population of potential solutions; they have some selection process based on fitness of individuals and some genetic
operators. One such system is a class of evolution strategies
(i.e., algorithms which imitate the principles of natural evolution
for parameter optimization problems). Fogels evolutionary pro-

1202

gramming is a technique for searching through a space of small

finite-state machines. Glovers scatter search techniques maintain a population of reference points and generate offspring by
weighted linear combinations. Another type of evolution based
systems is Hollands genetic algorithms [39], [97], [115]. In
1990, Koza proposed an evolution-based system, genetic programming, to search for the most fit computer program to solve
a particular problem [43], [140], [147].
Sheble et al. [39], applied the genetic algorithm to the
UC problem for one to seven days. The feasibility of genetic
algorithms application for UC problems has been examined for
both small- and large-size problems [27]. Maifeld et al. [132]
presented a new UC scheduling algorithm using genetic algorithm with domain-specific mutation operators. The robustness
of the proposed algorithm is demonstrated by comparison to
a Lagrangian Relaxation UC algorithm on different utilities.
Through a constraint handling technique, Yang et al. [55]
proposed a parallel genetic algorithm approach to solve UC
problem. Rudolf et al. [8] proposed a genetic algorithm for
solving the UC problem of a hydrothermal power system
and was tested on a real-scaled hydrothermal power system
over a period of a day in half-hour time steps for different
genetic algorithm parameters. Swarup et al. [71] employed
a new strategy for representing chromosomes and encoding
the problem search space, which is efficient and can handle
large-scale UC.
N. Evolutionary Programming
Yang et al. [51] and Juste et al. [69] presented an evolutionary
programming approach to UC problem and in which populations of contending solutions evolve through random changes,
competition, and selection. Chen et al. [53] presented a cooperative coevolutionary algorithm for UC. The proposed algorithm is an extension of the traditional evolutionary programming [100] which appears to have considerable potential for
formulating and solving more complex problems by explicitly
modeling the coevolution of cooperating spices.
O. Ant Colony Search Algorithm
The natural metaphor on which ant algorithms is based
is that of ant colonies. Real ants are capable of finding the
shortest path from a food source to their nest without using
visual cues by exploiting pheromone information. While
walking ants deposit pheromone on the ground and follow
in probability, pheromone was previously deposited by other
ants. The above behavior of real ants has inspired ant system.
This is an algorithm in which a set of artificial ants cooperate
to the solution of a problem by exchanging information via
pheromone deposited on graph edges. The ant system has
been applied to combinatorial optimization problems such as
UC problem. In [87], Sisworahardjo and El-Kaib have solved
UC using ant colony search algorithm. Similarly, in [124],
Huang has solved hydroelectric generation scheduling using
ant colony system-based optimization approaches.

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 2, MAY 2004

P. Hybrid Models
From the literature review, it has been observed that there exists a need for further improvement of the existing UC algorithm. In one end, an algorithm may be simple but suboptimal
and in the other end, complex but accurate [95], [96], [101],
[116]. So to achieve further improvement over the existing algorithm, one must complement the other (techniques). So mainly
more than one algorithm has been merged together and forms a
hybrid model to meet the industry requirement [56], [90], [144].
Su et al. [25] proposed a new fuzzy dynamic programming for
UC problem. A characteristic feature of the presented approach
is that the errors in the forecasted hourly loads can be taken into
account by using fuzzy set notations, making the approach superior to the conventional dynamic programming method. Similarly, Ouyang et al. [54] studied a hybrid ANN-dynamic programming approach to UCt. The proposed algorithm uses an
ANN to generate a preschedule according to the input load profile in the first step. Then, a dynamic search is performed at those
stages where the commitment states of some of the units are not
certain. Ouyang et al. [143] employed a multistage neural network-expert system approach to achieve real-time processing
results. The neural networks are used at the preprocessor and
postprocessor stages and the operating constraints are presented
as heuristic rules in the system where a feasible solution is obtained through inference. A new approach using genetic algorithms-based neural networks and dynamic programming has
been proposed by Huang et al. [125]. Mantawy et al. [3] integrated genetic algorithm, tabu search, and simulated annealing.
In [92] and N. P. Padhy integrated expert system, fuzzy system,
and neural network for a practical UC solution.
VI. CONCLUSION
This paper gives an overview of the concept of UC problem,
with a bibliographical survey of relevant background, practical
requirements, the historical events, the present state, and techniques. The citations listed in this bibliography provide a representative sample of current engineering thinking pertaining to
the next generation UC problem. A clear consensus is presently
heading toward the hybrid models, which is the combination
of both classical and nonclassical methods, and can handle the
present day complex UC problem commonly seen within developed countries. The genetic algorithms, evolutionary programming, and ant colony optimization technique are not thoroughly
tested yet. This paper is based on many research articles published in the last 30 years and periodic bibliographic updates on
this topic will be useful as the industry continues to evolve.
REFERENCES
[1] A. H. Mantawy, Y. L. Abdel-Magid, and S. Z. Selim, Unit commitment
by Tabu search, Proc. Inst. Elect. Eng., Gen. Transm. Dist., vol. 145,
no. 1, pp. 5664, Jan. 1998.
[2]
, A simulated annealing algorithm for unit commitment, IEEE
Trans. Power Syst., vol. 13, pp. 197204, Feb. 1998.
[3]
, Integrating genetic algoritms, Tabu search and simulated
annealing for the unit commitment problem, IEEE Trans. Power Syst.,
vol. 14, pp. 829836, Aug. 1999.
[4] A. I. Cohen, V. Brandwajn, and S. K. Chang, Security constrained unit
commitment for open market, in Proc. 21st IEEE Power Eng. Soc. Int.
Conf. Power Ind. Comput. Applicat., 1999, pp. 3944.

PADHY: UNIT COMMITMENTA BIBLIOGRAPHICAL SURVEY

[5] A. J. Wood and B. F. Wollenberg, Power Generation, Operation and

Control. New York: Wiley, 1996.
[6] A. L. Motto and F. D. Galiana, Equilibrium of auction markets with
unit commitment: The need for augmented pricing, IEEE Trans. Power
Syst., vol. 17, pp. 798805, Aug. 2002.
[7] A. Merlin and P. Sandrin, A new method for unit commitment at Electricite De France, IEEE Trans. Power App. Syst., vol. PAS-102, pp.
12181225, Aug. 1983.
[8] A. Rudolf and R. Bayrleithner, A genetic algorithm for solving the unit
commitment problem of a hydro-thermal power systems, IEEE Trans.
Power Syst., vol. 14, pp. 14601468, Nov. 1999.
[9] A. I. Cohen and S. H. Wan, A method for solving the fuel constrained
unit commitment problem, IEEE Trans. Power Syst., vol. PWRS-2, pp.
608614, Aug. 1987.
[10] A. I. Cohen and M. Yoshimura, A branch- and -bound algorithm for
unit commitment, IEEE Trans. Power App. Syst., vol. PAS-102, pp.
444451, Feb. 1983.
[11] A. I. Cohen and V. R. Sherkat, Optimization-based methods for operations, Proc. IEEE, vol. 77, pp. 15741590, Dec. 1987.
[12] B. F. Hobbs, S. Jitprapaikulsarn, S. Konda, U. Chan-kong, K. A. Loparo,
and D. J. Maratukulam, Analysis of the value for unit commitment
of improved load forecasts, IEEE Trans. Power Syst., vol. 14, pp.
13421348, Nov. 1999.
[13] C. K. Pang and H. C. Chen, Optimal short-term thermal unit commitment, IEEE Trans. Power App. Syst., vol. PAS-95, pp. 13361346,
July/Aug. 1976.
[14] C. K. Pang, G. B. Sheble, and F. Albu, Evaluation of dynamic programming based methods and multiple area representation for thermal
unit commitments, IEEE Trans. Power App. Syst., vol. PAS-100, pp.
12121218, Mar. 1981.
[15] C. Li, R. B. Johnson, and A. F. Svaboda, A new unit commitment
method, IEEE Trans. Power Syst., vol. 12, pp. 113119, Feb. 1997.
[16] C. Li, E. Hsu, A. J. Svoboda, C. Tseng, and R. B. Johnson, Hydro unit
commitment in hydro thermal optimization, IEEE Trans. Power Syst.,
vol. 12, pp. 764769, May 1997.
[17] C. Wang and S. M. Shahidehpour, Ramp-rate limits in unit commitment
and economic dispatch incorporating rotor fatigue effect, IEEE Trans.
Power Syst., vol. 9, pp. 15391545, Aug. 1994.
, Effects of ramp-rate limits on unit commitment and economic
[18]
dispatch, IEEE Trans. Power Syst., vol. 8, pp. 13411350, Aug. 1993.
[19] C. L. Chen and S. L. Chen, Short-term unit commitment with simplified
economic dispatch, Elect. Power Syst. Res., pp. 115120, June 1991.
[20] C. Li, E. Hsu, A. J. Svoboda, C. Tseng, and R. B. Johnson, Hydro unit
commitment in hydro-thermal optimization, IEEE Trans. Power Syst.,
vol. 12, pp. 764769, May 1997.
[21] C. Li, R. B. Johnson, and A. J. Svoboda, A new unit commitment
method, IEEE Trans. Power Syst., vol. 12, pp. 113119, Feb. 1997.
[22] C. Li, R. B. Johnson, A. J. Svoboda, C. Tseng, and E. Hsu, A robust unit
commitment algorithm for hydro-thermal optimization, IEEE Trans.
Power Syst., vol. 13, pp. 10511056, Aug. 1998.
[23] C. W. Richter Jr. and G. B. Sheble, A profit-based unit commitment
for the competitive environment, IEEE Trans. Power Syst., vol. 15, pp.
715721, May 2000.
[24] C.-P. Cheng, C.-W. Liu, and C.-C. Liu, Unit commitment by Lagrangian Relaxation and genetic algorithm, IEEE Trans. Power Syst.,
vol. 15, pp. 707714, May 2000.
[25] C.-C. Su and Y.-Y. Hsu, Fuzzy dynamic programming: an application
to unit commitment, IEEE Trans. Power Syst., vol. 6, pp. 12311237,
Aug. 1991.
[26] D. Chattopadhyay and J. Momoh, A multiobjective operations planning
model with unit commitment and transmission constraints, IEEE Trans.
Power Syst., vol. 14, pp. 10781084, Aug. 1999.
[27] D. Dasgupta and D. R. McGregor, Thermal unit commitment using genetic algorithms, Proc. Inst. Elect. Eng., Gen. Transm. Dist., vol. 141,
pp. 459465, Sept. 1994.
[28] D. Zhai, A. M. Breipohl, F. N. Lee, and R. Adapa, The effect of load
uncertainity on unit commitment risk, IEEE Trans. Power Syst., vol. 9,
pp. 510517, Feb. 1994.
[29] E. Castillo and E. Alvarez, Expert Systems: Uncertainity and Learning,
U.K.: Computational Mechanics Publications, 1991.
[30] E. Handschin and E. Slomski, Unit commitment in thermal systems
with long-term energy constraints, IEEE Trans. Power Syst., vol. 5, pp.
14701477, Nov. 1990.
[31] E. Tanlapco, J. Lawarree, and C. C. Liu, Hedging with future contracts
in a deregulated electricity industry, IEEE Trans. Power Syst., vol. 17,
pp. 577582, Aug. 2002.

1203

[32] F. Zhuang and F. D. Galiana, Toward a more rigorous and practical unit
commitment by Lagrangian Relaxation, IEEE Trans. Power Syst., vol.
3, pp. 763773, May 1988.
[33] F. N. Lee, J. Huang, and R. Adapa, Multi-area unit commitment via
sequential method and a dc power flow network model, IEEE Trans.
Power Syst., vol. 9, pp. 279287, Feb. 1994.
[34] F. N. Lee and Q. Feng, Multi-area unit commitment, IEEE Trans.
Power Syst., vol. 7, pp. 591599, May 1992.
[35] F. N. Lee, A fuel-constrained unit commitment method, IEEE Trans.
Power Syst., vol. 4, pp. 12081218, Aug. 1989.
[36]
, The application of commitment utilization factor (CUF) to
thermal unit commitment, IEEE Trans. Power Syst., vol. 6, pp.
691698, May 1991.
[37] F. N. Lee and Q. Chen, Unit commitment risk with sequential
rescheduling, IEEE Trans. Power Syst., vol. 6, pp. 10171023, Aug.
1991.
[38] F. N. Lee, Short-term unit commitmenta new method, IEEE Trans.
Power Syst., vol. 3, pp. 421428, May 1988.
[39] G. B. Sheble et al., Unit commitment by genetic algorithm with penalty
methods and a comparison of Lagrangian search and genetic algorithmeconomic dispatch example, Int. J. Elect. Power Energy Syst., vol. 18,
no. 6, pp. 339346, Feb. 1996.
[40] G. L. Kusic and H. A. Putnam, Dispatch and unit commitment including commonly owned units, IEEE Trans. Power App. Syst., vol.
PAS-104, pp. 24082412, Sept. 1985.
[41] G. Raina, S. Tong, and S. Nuelk, Unit commitment application builds
up existing EMS, IEEE Comput. Appl. Power, vol. 6, pp. 3539, Oct.
1993.
[42] G. S. Lauer, N. R. Sandell Jr., N. R. Bertsekas, and T. A. Posbergh, Solution Of large scale optimal unit commitment problems, IEEE Trans.
Power App. Syst., vol. PAS-101, pp. 7996, Jan. 1982.
[43] G. F. Luger and W. A. Stubblefield, Artificial Intelligence. Redwood
City, CA: Benjamin Cummings, 1993.
[44] G. B. Sheble and G. N. Fahd, Unit commitment literature synopsis,
IEEE Trans. Power Syst., vol. 9, pp. 128135, Feb. 1994.
[45] G. B. Sheble and T. T. Maifeld, Unit commitment by genetic algorithm
and expert system, Elect. Power Syst. Res., pp. 115121, Aug. 1994.
[46] H. H. Happ, R. C. Johnson, and W. J. Wright, Large scale hydro-thermal
unit commitment-method and results, IEEE Trans. Power App. Syst.,
vol. PAS-90, pp. 13731384, Aug. 1971.
[47] H. Ma and S. M. Shahidehpour, Decomposition approach to unit commitment with reactive constraints, Proc. Inst. Elect. Eng., Gen. Transm.
Dist., vol. 144, no. 2, pp. 113117, Mar. 1997.
[48] H. Mori and O. Matsuzaki, Embedding the priority into Tabu search for
unit commitment, in Proc. IEEE Winter Meeting, 2000.
[49] H. Sasaki, M. Watanabe, and R. Yokoyama, A solution method of unit
commitment by artificial neural networks, IEEE Trans. Power Syst.,
vol. 7, pp. 974981, Aug. 1992.
[50] H. Sasaki, M. Watanabe, J. Kubokawa, N. Yorina, and R. Yokoyama,
A solution method of unit commitment by artificial neural network,
in Proc. IEEE Power Eng. Soc. Summer Meeting, 1991.
[51] H. T. Yang, P. C. Yang, and C. L. Huang, Evolutionary programming
based economic dispatch for units with non smooth fuel cost functions,
IEEE Trans. Power Syst., vol. 11, pp. 112118, Feb. 1996.
[52] H. Ma and S. M. Shahidehpour, Unit commitment with transmission
security and voltage constraints, IEEE Trans. Power Syst., vol. 14, pp.
757764, May 1999.
[53] H. Chen and X. Wang, Cooperative coevolutionary algorithm for unit
commitment, IEEE Trans. Power Syst., vol. 16, pp. 128133, Feb. 2002.
[54] X. Hindsberger and H. F. M. Ravn, Multiresolution modeling of hydrothermal systems, in Proc. 22nd IEEE Power Eng. Soc. Int. Conf. Power
Ind. Comput. Applicat., 2001, pp. 510.
[55] H. Yang, P. Yang, and C. Huang, A parallel genetic algorithm approach
to solving the unit commitment problem: implementation on the transputer networks, IEEE Trans. Power Syst., vol. 12, pp. 661668, May
1997.
[56] H. Li, P. Chen, and H. Huang, Fuzzy Neural Intelligence Systems. Boca
Raton, FL: CRC, 2001.
[57] J. D. Guy, Security constrained unit commitment, IEEE Trans. Power
App. Syst., vol. PAS-90, pp. 13851389, May/June 1971.
[58] J. M. Arroyo and A. J. Conejo, A parallel repair genetic algorithm to
solve the unit commitment problem, IEEE Trans. Power Syst., vol. 17,
pp. 12161224, Nov. 2002.
[59] J. R. McDonald, G. M. Burt, J. S. Zielinski, and S. D. J. McArthur,
Intelligent Knowledge Based Systems in Electrical Power Engineering,
1st ed. London, U.K.: Chapman & Hall, 1997.

1204

[60] J. Valenzuela and M. Mazumdar, Making unit commitment decisions

when electricity is traded at spot market prices, in Proc. IEEE Winter
Meeting, 2001.
[61] J. J. Shaw, A direct method for security-constrained unit commitment,
IEEE Trans. Power Syst., vol. 10, pp. 13291342, Aug. 1995.
[62] K. Aoki, M. Itoh, T. Satoh, K. Nara, and M. Kanezashi, Optimal
long-term unit commitment in large scale systems including fuel
constrained thermal and pumped-storage hydro, IEEE Trans. Power
Syst., vol. 4, pp. 10651073, Aug. 1989.
[63] K. Aoki, T. Satoh, and M. Itoh, Unit commitment in large scale power
systems including fuel constrained thermal and pumped storage hydro,
IEEE Trans. Power Syst., vol. PWRS-2, pp. 10771084, May 1987.
[64] K. Ayoub and A. D. Patton, Optimal thermal generating unit commitment, IEEE Trans. Power App. Syst., vol. PAS-90, pp. 17521756,
July/Aug. 1971.
[65] K. D. Lee, J. T. Day, B. L. Cooper, and E. W. Gibbons, A global optimization method for scheduling thermal generation, hydro generation,
and economy purchase, IEEE Trans. Power App. Syst., vol. PAS-102,
pp. 19831986, July 1983.
[66] K. Hara, M. Kimura, and N. Honda, A method for planning economic
unit commitment and maintenance of thermal power systems, IEEE
Trans. Power App. Syst., vol. PAS-85, pp. 427436, May 1966.
[67] K. P. Wong and K. Doan, Artificial intelligence algorithm for daily
scheduling of thermal generators, Proc. Inst. Elect. Eng., Gen. Transm.
Dist., vol. 138, no. 6, pp. 518534, 1991.
[68] K. P. Wong and Y. W. Wong, Short term hydro thermal scheduling part
I: simmulated annealing approach, Proc. Inst. Elect. Eng., Gen. Transm.
Dist., vol. 141, pp. 497501, 1994.
[69] K. A. Juste, H. Kita, E. Tanaka, and J. Hasegawa, An evolutionary
programming solution to the unit commitment problem, IEEE Trans.
Power Syst., vol. 14, pp. 14521459, Nov. 1999.
[70] K. D. Lee, R. H. Vierra, G. D. Nagel, and R. T. Jenkins, Problems
associated with unit commitment in uncertainty, IEEE Trans. Power
App. Syst., vol. PAS-104, pp. 20722078, Aug. 1985.
[71] K. S. Swarup and S. Yamashiro, Unit commitment solution methodology using genetic algorithm, IEEE Trans. Power Syst., vol. 17, pp.
8791, Feb. 2002.
[72] K. Warwick, A. Ekwue, and R. Aggarwal, Artificial intelligence techniques in power systems, Inst. Elect. Eng. Power Eng. Series, 1997.
[73] K. Hussain, Solution method for unit commitment limitations and
utility constraints, IEEE Comput. Appl. Power, vol. 4, pp. 1620, Jan.
1991.
[74] K.-Y. Huang, H.-T. Yang, and C.-L. Huang, A new thermal unit commitment approach using constraint logic programming, IEEE Trans.
Power Syst., vol. 13, pp. 936945, Aug. 1998.
[75] L. F. B. Baptistella and J. C. Geromel, A decomposition approach to
problem of unit commitment schedule for hydro thermal systems, Proc.
Inst. Elect. Eng., Gen. Transm. Dist., vol. 127, no. 6, pp. 250258, 1980.
[76] M. E. Khan and R. Billinton, Generating unit commitment in composite generation and transmission systems, Proc. Inst. Elect. Eng.,
Gen. Transm. Dist., vol. 140, pp. 404410, Sept. 1993.
[77] M. Fotuhi-Firuzabad and R. Billinton, Unit commitment health
analysis in composite generation and transmission systems considering
stand-by units, Proc. Inst. Elect. Eng., Gen. Transm. Dist., vol. 146,
pp. 164168, Mar. 1999.
[78] M. Madrigal and V. H. Quintana, Existence and determination of competitive equilibrium in unit commitment power pool auctions, in Proc.
22nd IEEE Power Eng. Soc. Int. Conf. Power Ind. Comput. Applicat.,
2001, pp. 253257.
[79] M. Mazumdar and A. Kapoor, Stochastic models for power generation
system production costs, Elect. Power Syst. Res., vol. 35, pp. 93100,
1995.
[80] M. P. Walsh and M. J. O. Malley, Augmented Hopfield network for unit
commitment and economic dispatch, IEEE Trans. Power Syst., vol. 12,
pp. 17651774, Nov. 1997.
[81] M. Madrigal and V. H. Quintana, An interior-point/cutting-plane
method to solve unit commitment problems, IEEE Trans. Power Syst.,
vol. 15, pp. 10221027, Aug. 2000.
, Existence and determination of competitive equilibrium in unit
[82]
commitment power pool auctions: price setting and scheduling alternatives, IEEE Trans. Power Syst., vol. 16, pp. 380388, Aug. 2001.
[83] M.-S. Salam, A.-R.Abdul-Razak Hamdan, and K. M.Khalid Mohamed
Nor, Integrating an expert system into a thermal unit-commitment algorithm, Proc. Inst. Elect. Eng. C, vol. 138, pp. 553559, Nov. 1991.
[84] N. Chowdhury and R. Billinton, Unit commitment in interconnected
generating systems using a probabilistic technique, IEEE Trans. Power
Syst., vol. 5, pp. 12311238, Nov. 1990.

IEEE TRANSACTIONS ON POWER SYSTEMS, VOL. 19, NO. 2, MAY 2004

[85] N. Misra and Y. Baghzouz, Implementation of the unit commitment

problem on supercomputers, IEEE Trans. Power Syst., vol. 9, pp.
305310, Feb. 1994.
[86] N. R. Jimenez and A. J. Conejo, Short-term hydro thermal coordination
by Lagrangian Relaxation: Solution to the dual problem, in Proc. IEEE
Power Eng. Soc. Winter Meeting, 1998.
[87] N. S. Sisworahardjo and A. A. El-Kaib, Unit commitment using ant
colony search algorithm, in Proc. 2002 Large Eng. Syst. Conf. Power
Eng., 2002, pp. 26.
[88] N. P. Padhy, A new fuzzy expert decision making approach for unit
commitment with reliable risk reserve and emission constraints, J. Energy Environment, vol. 1, pp. 2536, Nov. 1999.
[89] N. P. Padhy and S. R. Paranjothi, Application of expert system for
short-term unit commitment problem, in Proc. Nat. Syst. Conf., Agra,
India, Jan. 1995, pp. 443447.
[90] N. P. Padhy, Hybrid Models for Unit Commitment Problems, Ph.D.
dissertation, Anna Univ., Chennai, India, 1997.
, Machine learning classifier system for unit commitment
[91]
problem, Int. J. Power Energy Syst., vol. 23, no. 1, pp. 4961, 2003.
, Unit commitment using hybrid models: a comparative study for
[92]
dynamic programming, expert system, fuzzy system and genetic algorithms, Elect. Power Energy Syst., vol. 23, pp. 827836, 2000.
[93] N. P. Padhy, V. R. Ramachandran, and S. R. Paranjothi, Fuzzy decision
system for unit commitment risk analysis, Int. J. Power Energy Syst.,
vol. 19, no. 2, pp. 180185, 1999.
, Fuzzy Decision System for Short-Term Unit Commitment
[94]
Problem, in Proc. Nat. Conf. Neural Networks Fuzzy Syst., Chennai,
India, March 1995, pp. 287298.
[95]
, Validation and verification of fuzzy-expert system : an application to short-term unit commitment problem, in Proc. Int. Conf. Elect.
Eng., vol. 2, China, Aug. 1996, pp. 10701073.
[96] N. P. Padhy, V. Ramachandran, and S. R. Paranjothi, A hybrid fuzzyneural network expert system for short-term unit commitment problem,
Int. J. Microelectronics Reliability, vol. 37, no. 5, pp. 733737, 1997.
[97] P. C. Yang, H. T. Yang, and C. L. Huang, Solving the unit commitment
problem with a genetic algorithm through a constraint satisfaction technique, Elect. Power Syst. Res., vol. 37, pp. 5565, 1996.
[98] P. G. Lowery, Generating unit commitment by dynamic programming,
IEEE Trans. Power App. Syst., vol. PAS-85, pp. 422426, May 1966.
[99] P. R. MacGregor and H. B. Puttgen, The integration of nonutility generation and spot prices within utility generation scheduling, in Proc.
IEEE Power Eng. Soc. Summer Meeting, 1994.
[100] P. Venkatesh, R. Gnanadass, and N. P. Padhy, Comparision and application of evolutionary programming techniques to combined economic
emission dispatch with line flow constraints, IEEE Trans. Power Syst.,
vol. 18, pp. 688692, May 2003.
[101] P. P. J Van Den Bosch and G. Honderd, A solution of the unit commitment problem via decomposition and dynamic programming, IEEE
Trans. Power App. Syst., vol. PAS-104, pp. 16841690, July 1985.
[102] Q. Zhai, X. Guan, and J. Cui, Unit commitment with identical units:
successive subproblem solving method based on Lagrangian Relaxation, IEEE Trans. Power Syst., vol. 17, pp. 12501257, Nov. 2002.
[103] R. Ferrero and S. M. Shahidehpour, Effect of deregulation on hydro
thermal system with transmission constraints, Elect. Power Syst. Res.,
vol. 38, pp. 191197, 1997.
[104] R. H. Kerr, J. L. Scheidt, A. J. Fontana, and J. K. Wiley, Unit commitment, IEEE Trans. Power App. Syst., vol. PAS-85, pp. 417421, May
1966.
[105] R. M. Burns and C. A. Gibson, Optimization of priority lists for a
unit commitment program, in Proc. IEEE Power Eng. Soc. Summer
Meeting, 1975.
[106] R. Nieva, A. Inda, and I. Guillen, Lagrangian reduction of search-range
for large-scale unit commitment, IEEE Trans. Power Syst., vol.
PWRS-2, pp. 465473, May 1987.
[107] R. Nieva, A. Inda, and J. Frausto, CHT: a digital computer package
for solving short term hydro-thermal coordination and unit commitment
problems, IEEE Trans. Power App. Syst., vol. PAS-1, pp. 168174,
Aug. 1986.
[108] R. H. Liang and F. C. Kang, Thermal generating unit commitment using
an extended mean field annealing neural network, Proc. Inst. Elect.
Eng., Gen. Transm. Dist., vol. 147, pp. 164170, May 2000.
[109] R. R. Shoults, S. K. Chang, S. Helmick, and W. M. Grady, A practical
approach to unit commitment, economic dispatch and savings allocation
for multiple-area pool operation with import/export constraints, IEEE
Trans. Power App. Syst., vol. PAS-99, pp. 625635, Mar./Apr. 1980.
[110] R. Baldick, The generalized unit commitment problem, IEEE Trans.
Power Syst., vol. 10, pp. 465475, Feb. 1995.

PADHY: UNIT COMMITMENTA BIBLIOGRAPHICAL SURVEY

[111] R. Billinton, M. Fotuhi-Firuzabad, and S. Aboreshaid, Unit commitment health analysis for interconnected systems, IEEE Trans. Power
Syst., vol. 12, pp. 11941201, Aug. 1997.
[112] S. K. Tong and S. M. Shahidehpour, Combination of Lagrangian-Relaxation and linear-programming approaches for fuel-constrained unit
commitment problems, Proc. Inst. Elect. Eng., Gen. Transm. Dist., vol.
136, pp. 162174, May 1989.
[113] S. Li, S. M. Shahidehpour, and C. Wang, Promoting the application
of expert systems in short-term unit commitment, IEEE Trans. Power
Syst., vol. 3, pp. 286292, Mar. 1993.
[114] S. Mukhtari, J. Singh, and B. Wollenberg, A unit commitment expert
system, IEEE Trans. Power Syst., vol. 3, pp. 272277, Feb. 1988.
[115] S. O. Orero and M. R. Irving, A genetic algorithm for generator scheduling in power systems, Elect. Power Energy Syst., vol. 18, no. 1, pp.
1926, 1996.
, Large scale unit commitment using a hybrid genetic algorithms,
[116]
Int. J. Elect. Power Energy Syst., vol. 19, no. 1, pp. 4555, 1997.
[117] S. Sen and D. P. Kothari, Optimal thermal generating unit commitment:
a review, Int. J. Elect. Power Energy Syst., vol. 20, no. 7, pp. 443451,
1998.
[118] S. Vemuri and L. Lemonidis, Fuel constrained unit commitment, IEEE
Trans. Power Syst., vol. 7, pp. 410415, Feb. 1991.
[119] S. K. Tong and S. M. Shahidehpour, Hydro thermal unit commitment
with probabilistic constraints using segmentation method, IEEE Trans.
Power Syst., vol. 5, pp. 276282, Feb. 1990.
[120] S. Takriti and J. R. Birge, Using integer programming to refine Lagrangian-based unit commitment solutions, IEEE Trans. Power Syst.,
vol. 15, pp. 151156, Feb. 2000.
[121] S. Takriti, J. R. Birge, and E. Long, A stochastic model for the unit commitment problem, IEEE Trans. Power Syst., vol. 11, pp. 14971508,
Aug. 1996.
[122] S. Saneifard, N. R. Prasad, and H. A. Smolleck, A fuzzy logic approach
to unit commitment, IEEE Trans. Power Syst., vol. 12, pp. 988995,
May 1997.
[123] S.-Y. Lai and R. Baldick, Unit commitment with ramp multipliers,
IEEE Trans. Power Syst., vol. 14, pp. 5864, Feb. 1999.
[124] S.-J. Huang, Enhancement of hydroelectric generation scheduling
using ant colony system based optimization approaches, IEEE Trans.
Energy Conversion, vol. 16, pp. 296301, Sept. 2001.
[125] S.-J. Huang and C.-L. Huang, Application of genetic-based neural network to thermal unit commitment, IEEE Trans. Power Syst., vol. 12,
pp. 654660, May 1997.
[126] S. Ruzic and N. Rajakovic, A new approach for solving extended unit
commitment problem, IEEE Trans. Power Syst., vol. 6, pp. 269277,
Feb. 1991.
[127] S. Sen and D. P. Kothari, Evaluation of benefit of inter-area energy
exchange of the indian power systems based on multi-area unit commitment approach, Elect. Mach. Power Syst., vol. 26, pp. 801813, 1998.
[128] S. Virmani, E. C. Adrian, K. Imhof, and S. Muhherjee, Implementation
of a Lagrangian based unit commitment problem, IEEE Trans. Power
Syst., vol. 4, pp. 13731380, Nov. 1989.
[129] T. J. Larsen, I. Wangensteen, and T. Gjengedal, Sequential time step
unit commitment, in Proc. IEEE Winter Meeting, 2001.
[130] T. S. Dillon, K. W. Edwin, H. D. Kochs, and R. J. Taud, Integer programming approach to the problem of optimal unit commitment with
probabilistic reserve determination, IEEE Trans. Power App. Syst., vol.
PAS-97, pp. 21542166, Nov./Dec. 1978.
[131] T. K. Siu, G. A. Nash, and Z. K. Shawwash, A practical hydro, dynamic
unit commitment and loading model, IEEE Trans. Power Syst., vol. 16,
pp. 301306, May. 2001.
[132] T. T. Maifeld and G. B. Sheble, Genetic-based unit commitment algorithm, IEEE Trans. Power Syst., vol. 11, pp. 13591370, Aug. 1996.
[133] U. D. Annakkage, T. Numnonda, and N. C. Pahalawaththa, Unit commitment by parallel simmulated annealing, Proc. Inst. Elect. Eng., Gen.
Transm. Dist., vol. 142, pp. 595600, 1995.
[134] Y. R. Sood, N. P. Padhy, and H. O. Gupta, Discussion on optimal power
flow by enhanced genetic algorithms, IEEE Trans. Power Syst., vol. 18,
p. 1219, Aug. 2003.

1205

[135] W. J. Hobbs, G. Hermon, S. Warner, and G. B. Sheble, An enhanced dynamic programming approach for unit commitment, IEEE Power Eng.
Rev., p. 70, Aug. 1988.
[136] W. L. Snyder, Jr., H. D. Powell, Jr., and J. C. Rayburn, Dynamic programming approach to unit commitment, IEEE Trans. Power Syst., vol.
2, pp. 339347, May 1987.
[137] X. Bai and S. M. Shahidehpour, Extended neighborhood search algorithm for constrained unit commitment, Int. J. Elect. Power Energy
Syst., vol. 19, no. 5, pp. 349356, 1997.
[138] X. Guan, P. B. Luh, and H. Yan, An optimization based method for unit
commitment, Int. J. Elect. Power Energy Syst., vol. 14, no. 1, pp. 917,
Feb. 1992.
[139] X. Lei, E. Lerch, and C. Y. Xie, Frequency security constrained
short-term unit commitment, Elect. Power Syst. Res., vol. 60, pp.
193200, Jan. 2002.
[140] X. Ma, A. A. El-Keib, R. E. Smith, and H. Ma, A genetic algorithm
based approach to thermal unit commitment of electric power systems,
Elect. Power Syst. Res., vol. 34, pp. 2936, 1995.
[141] Y.-Y. Hsu, C.-C. Su, C.-C. Liang, C.-J. Lin, and C.-T. Huang, Dynamic
security constrained multi-area unit commitment, IEEE Trans. Power
Syst., vol. 6, pp. 10491055, Aug. 1991.
[142] Z. Ouyang and S. M. Shahidehpour, Heuristic multi-area unit commitment with economic dispatch, Proc. Inst. Elect. Eng., Gen. Transm.
Dist., vol. 138, pp. 242252, May 1991.
[143]
, A hybrid artificial neural network-dynamic programming
approach to unit commitment, IEEE Trans. Power Syst., vol. 7, pp.
236242, Feb. 1992.
, A multi-stage intelligent system for unit commitment, IEEE
[144]
Trans. Power Syst., vol. 7, pp. 639646, May 1992.
[145]
, An intelligent dynamic programming for unit commitment application, IEEE Trans. Power Syst., vol. 6, pp. 12031209, Aug. 1991.
, Short-term unit commitment expert system, Elect. Power Syst.
[146]
Res., pp. 113, Dec. 1990.
[147] Z. Michalewicz, Genetic Algorithm+Data Structure=Evolution Programs, 2nd ed. New York: Springer-Verlag, 1992.
[148] A. Rajan, C. C. Mohan, and M. R. Manivannan, Neural based tabu
search method for solving unit commitment problem, in Proc. 5th Int.
Conf. Power Syst. Manage. Contr., 2002, pp. 180185.
[149] H. Mori and T. Usami, Unit commitment using tabu search with restricted neighborhood, in Proc. Intell. Syst. Applicat. Power Syst., 1996,
pp. 422427.
[150] W.-M. Lin, F.-S. Cheng, and M.-T. Tsay, An improved tabu search for
economic dispatch with multiple minima, IEEE Trans. Power Syst., vol.
17, pp. 108112, Feb. 2002.
[151] A. H. Mantawy, S. A. Soliman, and M. E. El-Hawary, A new tabu search
algorithm for the long-term hydro scheduling problem, in Proc. Large
Eng. Syst. Conf. Power Eng., June 2002, pp. 2934.

Narayana Prasad Padhy received the electrical engineering and M.E. degrees in power system engineering (Hons.) from Madurai Kamaraj University,
India, in 1990 and 1993, respectively. He received
the Ph.D. degree in electrical engineering from Anna
University, Chennai, India, in 1997.
Currently, he is an Assistant Professor in the
Department of Electrical Engineering at the Indian
Institute of Technology (IIT), Roorkee, India. He
was Assistant Professor in the Electrical Engineering
Department at the Birla Institute of Technology
and Science (BITS), Pilani, India, in 1997. He taught courses in basic
electrical engineering, power system, and artificial intelligence. His research
interests include power systems privatization, restructuring and deregulation,
artificial intelligence applications to power systems operation and optimization
problems, unit commitment, power systems wheeling, and FACTS.